mlsanityData to 7 Oct 2026

When Predictions Shape Reality: A Socio-Technical Synthesis of Performative Predictions in Machine Learning

arXiv 2601.04447 ↗First posted 7 Jan 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Machine learning models are increasingly used in high-stakes domains where their predictions can actively shape the environments in which they operate, a phenomenon known as performative prediction. This dynamic, in which the deployment of the model influences the very outcome it seeks to predict, can lead to unintended consequences, including feedback loops, performance issues, and significant societal risks. While the literature in the field has grown rapidly in recent years, a socio-technical synthesis that systemises the phenomenon concepts and provides practical guidance has been lacking. This Systematisation of Knowledge (SoK) addresses this gap by providing a comprehensive review of the literature on performative predictions. We provide an overview of the primary mechanisms through which performativity manifests, present a typology of associated risks, and survey the proposed solutions offered in the literature. Our primary contribution is the ``Performative Strength vs. Impact Matrix" assessment framework. This practical tool is designed to help practitioners assess the potential influence and severity of performativity on their deployed predictive models and select the appropriate level of algorithmic or human intervention.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 56 of its 81 references that are in the map. First posted on arXiv 7 Jan 2026.

Builds on

56 of 81 references.

References · 56

  1. 2020Performative Predictioncited by 318
  2. 2021Outside the Echo Chamber: Optimizing the Performative Riskcited by 95
  3. 2020Stochastic Optimization for Performative Predictioncited by 92
  4. 2020Performative Prediction in a Stateful Worldcited by 90
  5. 2021How to Learn when Data Reacts to Your Model: Performative Gradient Descentcited by 70
  6. 2020Stochastic optimization with decision-dependent distributionscited by 65
  7. 2024Fairness Feedback Loops: Training on Synthetic Data Amplifies Biascited by 55
  8. 2022Multiplayer Performative Prediction: Learning in Decision-Dependent Gamescited by 42
  9. 2022Performative Powercited by 42
  10. 2022Regret Minimization with Performative Feedbackcited by 40
  11. 2021State Dependent Performative Prediction with Stochastic Approximationcited by 38
  12. 2022Anticipating Performativity by Predicting from Predictionscited by 37
38 more
  1. 2022Multi-agent Performative Prediction: From Global Stability and Optimality to Chaoscited by 33
  2. 2023Performative Prediction with Neural Networkscited by 33
  3. 2022Making Decisions under Outcome Performativitycited by 32
  4. 2022Decision-Dependent Risk Minimization in Geometrically Decaying Dynamic Environmentscited by 31
  5. 2021How to Learn when Data Gradually Reacts to Your Modelcited by 29
  6. 2022Performative Reinforcement Learningcited by 25
  7. 2022Multi-agent Performative Prediction with Greedy Deployment and Consensus Seeking Agentscited by 23
  8. 2023Plug-in Performative Optimizationcited by 20
  9. 2023Performative Federated Learning: A Solution to Model-Dependent and Heterogeneous Distribution Shiftscited by 14
  10. 2024Performative Reinforcement Learning in Gradually Shifting Environmentscited by 13
  11. 2022Stochastic Approximation with Decision-Dependent Distributions: Asymptotic Normality and Optimalitycited by 12
  12. 2023A Classification of Feedback Loops and Their Relation to Biases in Automated Decision-Making Systemscited by 12
  13. 2023When accurate prediction models yield harmful self-fulfilling propheciescited by 12
  14. 2024Stochastic Optimization Schemes for Performative Prediction with Nonconvex Losscited by 12
  15. 2021Approximate Regions of Attraction in Learning with Decision-Dependent Distributionscited by 11
  16. 2022Fair When Trained, Unfair When Deployed: Observable Fairness Measures are Unstable in Performative Prediction Settingscited by 11
  17. 2023Performative Prediction with Bandit Feedback: Learning through Reparameterizationcited by 11
  18. 2024Addressing Polarization and Unfairness in Performative Predictioncited by 11
  19. 2023Zero-Regret Performative Prediction Under Inequality Constraintscited by 10
  20. 2024Optimal Classification under Performative Distribution Shiftcited by 9
  21. 2023Two-timescale Derivative Free Optimization for Performative Prediction with Markovian Datacited by 8
  22. 2024Performative Reinforcement Learning with Linear Markov Decision Processcited by 8
  23. 2024Distributionally Robust Performative Predictioncited by 7
  24. 2024Learning In Reverse Causal Strategic Environments With Ramifications on Two Sided Marketscited by 6
  25. 2024Algorithmic Fairness in Performative Policy Learning: Escaping the Impossibility of Group Fairnesscited by 6
  26. 2024Learning the Distribution Map in Reverse Causal Performative Predictioncited by 5
  27. 2025Decision-Dependent Stochastic Optimization: The Role of Distribution Dynamicscited by 5
  28. 2022Prisoners of Their Own Devices: How Models Induce Data Bias in Performative Predictioncited by 4
  29. 2022Optimizing the Performative Risk under Weak Convexity Assumptionscited by 4
  30. 2024Clipped SGD Algorithms for Performative Prediction: Tight Bounds for Clipping Bias and Remediescited by 4
  31. 2024The Limitations of Model Retraining in the Face of Performativitycited by 4
  32. 2024Tight Lower Bounds and Improved Convergence in Performative Predictioncited by 4
  33. 2021Bayesian logistic regression for online recalibration and revision of risk prediction models with performance guaranteescited by 3
  34. 2022Long Term Fairness for Minority Groups via Performative Distributionally Robust Optimizationcited by 3
  35. 2023Designing monitoring strategies for deployed machine learning algorithms: navigating performativity through a causal lenscited by 3
  36. 2024Evaluating and Correcting Performative Effects of Decision Support Systems via Causal Domain Shiftcited by 3
  37. 2024Practical Performative Policy Learning with Strategic Agentscited by 3
  38. 2024Identifying Predictions That Influence the Future: Detecting Performative Concept Drift in Data Streamscited by 3

50 most cited of 56 All 56 (JSON)

Built on by

Not cited yet.